Patent Application
20140022945
The invention pertains to communications networks that comprise at least one group of nodes connected to one another by oriented links, and more specifically the optimization of the use of such nodes, particularly in to terms of energy consumption.
Here, the term “oriented link” refers to a link, whether wired or wireless, capable of enabling the transmission of signals from a first node to a second node, and therefore in a chosen direction. Consequently, here the term “link” refers to a transmission, whether wired or wireless, capable of enabling the transmission of signals between two nodes, independent of the direction.
It should be noted that the invention relates to any type of communications network, including optical networks.
As is known to a person skilled in the art, certain communications networks, called “dynamic”, comprise nodes, generally called routers or switches, which are equipped with elements, such as line-cards, optical interfaces equipped with lasers, optical cross-connects (or OXCs), or electronic interfaces, which consume energy when they are placed in a state called “active” (or “up”) or in a state called “partially active” (or “idle”).
Here, the term “up state” refers to a state in which all the components of an element are ready to be used and therefore consume energy, and “idle state” refers to an intermediate state in which certain components of an element consume energy in order to be ready for immediate use, as it takes a relatively long time to switch them from the so-called off (or inactive or “down”) state to the active (or up) state, while some others are placed in the off (or down) state, and therefore consume no energy, as it takes a very short length of time to switch them from the off (or down) state to the active (or up) state.
Thanks to this type of multiple-operation element, it is possible to manage the energy consumption of a part of a (communications) network based on the needs of the moment. This management is generally provided by a network management system (or NMS). It should be understood that there is no need to temporarily feed some elements of a node in whole or in part, which are dedicated to an inter-node link if that link's maximum transmission capacity available is only partially used, or not at all used, temporarily.
Unfortunately, no routing protocol provides for the compilation of elements that have been placed by the network management system into an off (or down) state, or a partially on (or idle) state, which harms the routing within networks which are called “energy-efficient” and more precisely, the calculation of the routing paths, because all the elements of active nodes are then considered to be placed in their active (or on, or up) state. Additionally, this prohibits any actually dynamic adaptation of the nodes as a function of transmission capacity allocation demands.
The purpose of the invention is therefore to improve the situation.
According to a first aspect, the invention proposes a method, dedicated to the management of information related to the first operating state of a group of nodes of a communications network, linked to one another by oriented links Ljj′ each associated with a maximum transmission capacity controlled by elements of nodes that may be placed in a state from among P operating states (where P≧2), chosen from among an active state and an off state and/or at least one intermediate idle state, and comprising the following steps:
The method may comprise other characteristics that may be taken separately or in combination, and in particular:
According to a second aspect, the invention proposes a device, whether centralized or distributed, that first, is dedicated to the management of information related to the operating state of a group of nodes of a communications network, connected to one another by oriented links Ljj′ each associated with a maximum transmission capacity controlled by elements of nodes that may be placed into a state from among P operating states (where P≧2), chosen from among an on state and an off state and/or at least one intermediate idle state, and second, is capable of implementing a method of the type presented above.
According to a third aspect, the invention proposes a network equipment that firstly is capable of forming part of a communications network and of being connected to other nodes of that network by associated oriented links Ljj′ each associated with a maximum transmission capacity controlled by elements of nodes that may be placed in a state from among P operating states (where P≧2), chosen from among an up state and a down state and/or at least one intermediate idle state, and additionally comprising an information management device of the type presented above.
Other characteristics and advantages of the invention shall become apparent upon examining the detailed description below, and the attached drawings, in which:
The appended drawings might not merely serve to complete the invention, they might contribute to the definition of the invention, if necessary.
It is a particular object of the invention to make it possible to compile, with a routing protocol, operating states of the node elements (Nj) belonging to a communications network (RC) of the type known as “energy-efficient”, particularly for the purpose of optimizing the routing of the communications.
It is assumed hereinafter, by way of an example, that the communications network (RC) is an optical network. However, the invention is not limited to this sort of communications terminal. It pertains to any communications network comprising at least one group of nodes comprising elements that need energy to function and that may selectively and dynamically be placed in one state from among P operating states (where P≧2), chosen from among an on (or up) state and an off (or down) state and/or at least one partially on (or idle) state. It shall therefore be understood that the elements may feature either an up state and a down state, or an upstate and at least one idle state, or an up state, a down state, and at least one idle state.
Consequently, the invention particularly pertains to optical networks, particularly WSON (“Wavelength Switched Optical Network”), IP/MPLS (“Internet Protocol/MultiProtocol Label Switching”) Internet protocol and control plane networks, and Carrier Grade Ethernet Networks, and all other types of data transport networks.
Furthermore, it is assumed hereinafter, by way of an illustrative example, that the node elements can only assume three different operating states (i.e. the up state, the down state, and a single idle state). Consequently, the number P is equal to 3. However, this number P may be equal to 2 (particularly when there is no idle state) or greater than 3.
It is important to note that a link L′k established between two nodes Nj and Nj′ (j′≠j) comprises a first oriented link Ljj′ running from the node Nj to the node Nj′ and/or a second oriented link Lj′j running from the node Nj′ to the node Nj.
It should also be noted that in the example depicted in
It should also be noted that in the example depicted in
For example, the nodes Nj are routers or cross-connects.
It should also be noted that each node Nj comprises elements of the aforementioned type, associated with oriented links Ljj′ or Lj′j. In fact, each oriented link Ljj′ leaving a node Nj is coupled to one or more elements of that node Nj which are dedicated to it, and each oriented link Lj′j arriving at a node Nj is coupled to one or more elements of that node Nj.
For example, the elements are line-cards, optical interfaces equipped with lasers, optical cross-connects (or OXCs), or electronic interfaces.
It should also be noted that each oriented link Ljj′ has a predefined maximum transmission capacity CTMjj′. Generally, the two oriented links Ljj′ and Lj′j of the same link L′k have the same predefined maximum transmission capacity. However, the invention also applies to cases in which the two oriented links Ljj′ and Lj′j of a single link L′k respectively have different predefined maximum transmission capacities.
The invention proposes implementing within at least one group of nodes Nj of a network RC a method devoted to managing information related to the operating state of these nodes Nj.
This method comprises at least the two main steps (i) and (ii).
A first step (i) consists of obtaining, for each oriented link Ljj′ of each node Nj, P values Vjj′i, where i=1 to P, respectively representative of the distribution percentages of its maximum transmission capacity CTMjj′ between elements placed within these P different operating states.
For example, when P=3, i.e. whenever there is one defined idle state, for each oriented link Ljj′ the following are obtained:
A transmission capacity of an oriented link Ljj′ is said to be available when the elements of the node Nj that are associated with that oriented link Ljj′ are all placed in an up (or active) state. It is important to note that the available transmission capacity CTDjj′ of an oriented link Ljj′ is divided into two complementary parts: an available, used transmission capacity CTDAjj′ and an available, usable transmission capacity CTDBjj′ (where CTDjj′=CTDAjj′+CTDBjj′). This is because the available transmission capacity CTDjj′ of an oriented link Ljj′ is not necessarily all used at a given moment.
A transmission capacity of an oriented link Ljj′ is said to be partially available when the elements of the node Nj that are associated with that oriented link Ljj′ are partially placed in the up state (some are up and some are down).
A transmission capacity of an oriented link Ljj′ is said to be available when the elements of the node Nj that are associated with that oriented link Ljj′ are all placed in the down (or off) state.
Preferentially, these are the nodes Nj of a group that provide, during the first step (i) the values Vjj′i which concern them by means of messages. This providing may occur periodically, or upon a request, or whenever the state of one of the elements of a node Nj is modified.
It should be noted that the messages containing the values Vjj′ may be transmitted by each node Nj to the neighboring nodes Nj′ (j′≠j) to which it is connected by a link L′k, and/or to an inventive device D tasked with managing information that relate to the operating state of the nodes Nj of at least the group to which it belongs. It is important to note that the device D may work either in a centralized way for all the nodes of at least one group of a network RC, in which case it forms part of a network device ER as depicted in a non-exhaustive fashion in
By way of a non-exhaustive example, the messages may comply with a chosen link state routing protocol. For example, certain messages of the OSPF-TE (“Open Shortest Path First-Traffic Engineering”) protocol, which is well known to the person skilled in the art and is used to exchange information about traffic engineering, may be used. In this case, the P values Vjj′i of each node Nj may, for example, be integrated into P dedicated Type-Length-Value (or TLV), fields, potentially type 6, of Link State Advertisement (or LSA) messages of the OSPF-TE v.2 protocol.
Purely as an illustrative example, a new TLV subfield (or “sub-TLV”), for example called “Idle Bandwidth”, may define the idle capacity. This subfield may be type x, where x is a value four bytes long that had not yet been allocated by the IANA (“Internet Assigned Numbers Authority”).
Likewise, a new TLV subfield (or “sub-TLV”), for example called “Down Bandwidth”, may define the down capacity. This sub-field may be type y, where y is a value four bytes long that has not yet been allocated by the IANA.
The active capacity may be defined by the TLV subfield (or “sub-TLV”) called “Up Bandwidth”, which is an extension of the maximum capacity (or “Maximal Bandwidth”) defined by the RFC3630 rule of the IETF.
However, other messages of other routing protocols may be used to transmit the value Vjj′, and particularly certain messages of the standard IS-IS-TE protocol (for “Intermediate System-Intermediate System-Traffic Engineering) or any proprietary routing protocol that makes it possible to transport operating state information of the network equipment.
A second step (ii) of the inventive method consists of building with the obtained values Vjj′i an operating state table of the group's nodes Nj, which is representative for each of the oriented links Ljj′ of its available CTDjj′ and unavailable CTIjj′ transmission capacities, and, if they exist, partially available transmission capacities CTPjj′.
This device D is the one that is tasked with constituting this operating state table. It should be noted that the device D preferentially updates this table periodically, and not necessarily each time that it receives new value Vjj′. The update period is a configuration parameter of the routing protocol, which transports the operating state information.
In this case, this operating state table is stored by the device D in storage means MS′ and provided to all the nodes Nj of the group so that it can store them in the storage means MS. It should be noted that whenever a node Nj is connected to nodes belonging to multiple different groups, it is preferable for it to store within its storage means MS the different operating state tables of these different groups. It should also be noted that an operating state table may potentially be stored in a distributed fashion in multiple nodes Nj.
The storage means MS and MS′ may come in any form known to the person skilled in the art, and particularly in the form of a memory or database, potentially a software- or file-based one.
It should be noted that an operating state table may be constructed in a manner similar to that of a link state table, which is well-known to the person skilled in the art. In this case, the operating state table is accessible via a routing protocol engine, such as the same one, though adapted, as the one used for the OSPF-TE protocol.
It should also be noted that during the second step (ii), based on the table of operating states, at least P graphs G1 may also be constructed that are each representative of oriented links Ljj′ between the nodes Nj of a group and values Vjj′i associated with one of the P operating states. These graphs G1 shall particularly be useful during the phase of determining the routing path, which shall be discussed later on.
For example, when P=3, the following graphs may be drawn for a group of nodes Nj:
The inventive method may also and advantageously comprise a third step (iii) that is performed by the device D whenever it receives a request to allocate a transmission capacity CTR for a communication that must go through an input node Nj of the group and an output node Nj′ of that same group.
Whenever the aforementioned situation occurs, at least one inter-node routing path Cn that takes oriented links Ljj′ established between the input Nj and output Nj′ nodes is determined (for example, by the device D) based on the operating state table (stored within its storage means (MS′) and to on auxiliary information that relates to the actual (current) use of the available transmission capacities CTDjj′ of the oriented links Ljj′.
It should be noted that when the device D is implemented within a node Nj, it determines inter-node routing paths Cn for its own node Nj, and it may also, potentially, determine inter-node routing paths Cn′ for other nodes that request this from it.
The auxiliary information particularly comprises the available, used transmission capacity CTDAjj′ of an oriented link Ljj′ and/or the available, usable transmission capacity CTDBjj′ of that same oriented link Ljj′ (where CTDjj′=CTDAjj′+CTDBjj′). The information CTDjj′ comes, for example, from configuration information of the network infrastructure and traffic engineering information CTDAjj′ regarding the usage of the network.
In step (iii), every time that an oriented link Ljj′ has an available, usable transmission capacity CTDjj′ which is less than the transmission capacity CTR defined within a received allocation request, it may be determined whether part of the partially available transmission capacity CTPjj′ of that same link Ljj′ can be used as a complement of the available, usable transmission capacity CTDBjj′ to meet that received allocation request. If it can, that oriented link Ljj′ may be retained, whereas if it can't, that oriented link Ljj′ might not be retained.
In other words, if CTR<CTDBjj′ or CTR<CTDBjj′+CTPjj′, then the oriented link Ljj′ is retained for computing the routing paths Cn, and if CTR>CTDBjj′+CTPjj′, then the oriented link Ljj′ is not retained for computing the routing paths.
In one variant embodiment, in step (iii), every time that an oriented link Ljj′ has an available, usable transmission capacity CTDBjj′ which is less than the transmission capacity CTR defined within a received allocation request, it may be determined whether part of the partially available transmission capacity CTPjj′ of that same link Ljj′ can be used as a complement of the available, usable transmission capacity CTDBjj′ to fulfill that received allocation request. If it can, that oriented link Ljj′ may be retained, while if it can't, it may be determined whether part of the unavailable transmission capacity CTIjj′ of that same inter-node link may be used as a to complement to the available, usable transmission capacity CTDBjj′ to fulfill that received allocation request. Then, if it can, that oriented link Ljj′ may be retained, whereas if it can't, that oriented link Ljj′ might not be retained.
In other words, if CTR<CTDBjj′, or CTR<CTDBjj′+CTPjj′, or CTR<CTDBjj′+CTPjj′+CTIjj′, or CTR<CTDBjj′+CTIjj′, then the oriented link Ljj′ for computing the routing paths Cn, and if CTR>CTDBjj′+CTPjj′+CTIjj′, then the oriented link Ljj′ is not retained for computing the routing paths.
Once the oriented links Ljj′ that might participate in routing paths have been determined, at least one routing path Cn is computed. Preferentially, multiple routing paths Cn are computed, for example two or three, then one of them is selected based on at least one chosen criterion.
Any criterion known to the person skilled in the art that may serve to select a routing path from among more than one may be used here. Here, one may, for example, use a criterion that relates to the number of inter-node oriented links Ljj′ in the path, in which case preference is given to the shortest one (i.e. the one that has the fewest oriented links Ljj′), and/or an energy cost criterion, in which case preference is given to the routing path that consumes the least energy, and/or a location criterion, in which case preference is given to oriented links Ljj′ that pass through nodes situated in chosen locations, and/or a load balancing criterion, in which case preference is given to oriented links Ljj′ whose load rate is below a threshold. Other criteria and/or a combination of multiple criteria may also be used, including those mentioned above.
It should be noted that at least one of the criteria used during the path selection Cn may be adjustable as a function, for example, of the load rate of the oriented links Ljj′. In other words, weights may be assigned to each of the criteria used, and its weights may be varied based on the load rate of the oriented links Ljj′. This way, each network operator may set its own criteria with its own weights depending on its own network infrastructure configuration strategies.
If the selected routing path Cn requires an operating state change in one or more elements of one or more nodes, these nodes must be alerted in order for them to reconfigure. As a reminder, this situation occurs every time it has been determined that some of the partially available transmission capacity CTPjj′ and/or some of the unavailable transmission capacity CTIjj′ of at least one oriented link Ljj′ was/were necessary to complement the available, usable transmission capacity CTDBjj′ of that same oriented link Ljj′ to fulfill a received allocation request. In this case, during the third step (iii) a message is addressed to at least one of the two nodes Nj and Nj′ that are concerned by that oriented link Ljj′, ordering it to put at least one designated element associated with that oriented link Ljj′ into an up state, which had until then been placed in an idle or down state. This way, each oriented link Ljj′ of the selected routing path Cn will be able to fulfill the received allocation request.
In
It is assumed that the first path C1 is now capable of largely fulfilling the allocation request requiring a transmission capacity CTR equal to 5λ. This is because we have gone from an initial situation in which each oriented link Ljj′ of the first path C1 has CTDAjj′=14λ available and used and CTDBjj′=2λ available and usable, to an intermediate situation in which each oriented link Ljj′ of the first path C1 has 14λ available and used and 18λ available and usable after the state change, and then from the aforementioned intermediate situation to a final situation in which each oriented link Ljj′ of the first path C1 to has 19λ available and used and 13λ available and usable after the required 5λ have been allocated.
In
It should be understood that the second path C2 is capable of exactly fulfilling the allocation request requiring a transmission capacity CTR equal to 5λ, with no state change and after the required 5λ has been allocated. We have therefore gone from an initial situation in which the oriented link L13 of the second path C2 has 25λ available and used and 7λ available and usable, the oriented link L34 of the second path C2 has 34λ available and used and 14λ available and usable, the oriented link L45 of the second path C2 has 43λ available and used and 5λ available and usable, and the oriented link L56 of the second path C2 has 30λ available and used and 6λ available and usable, to a final situation in which the oriented link L13 of the second path C2 has 30λ available and used and 2λ available and usable, the oriented link L34 of the second path C2 has 39λ available and used and 9λ available and usable, the oriented link L45 of the second path C2 has 48λ available and used and 0λ available and usable, and the oriented link L56 of the second path C2 has 35λ available and used and 1λ available and usable, after the required 5λ has been allocated.
It should be understood that the first path C1 shall be selected when the criteria with the greatest weights are load-balancing and available and usable transmission capacity on each oriented link Ljj′, while the second path C2 shall be selected when the criteria with the greatest weight is energy cost (establishing the path C2 does not actually require any additional energy).
One illustrative and therefore non-exhaustive example of a path-computing algorithm is given in the appendix.
It should be noted that it may also be beneficial to have a transmission capacity allocation for a routing path in order to switch elements, which are temporarily not serving, from an active or partially active state to a passive state, in order to temporarily save energy.
It shall also be noted that the device D is preferentially constructed in the form of software (or computer) modules. However, it may also be constructed in the form of a combination of electronic circuits and software (or computer) modules.
The invention offers several advantages, which include:
The invention is not limited to the embodiments of the information management device, network equipment, and information management method described above, which are only given as examples; rather, it encompasses all variants that a person skilled in the art may envision within the scope of the claims below.
An example path-computing algorithm that makes it possible to preserve the consumed energy is given below. In this example, the variable “X” represents the length of a path Cn′ from a start node (router A) to its neighbor, node R, if that path Cn′ is to pass through a node R′. If that path Cn′ is shorter than the selected and saved current path Cn in reaching the node R, the current path Cn is replaced by that path Cn′ (associated with X). Furthermore, the variable “dist_entre (R′, R)” represents the length between to the two neighboring nodes R′ and R with the weighted graph links according to three operating states: active (or UP) partially active (or IDLE), and off (or DOWN).
| Number | Date | Country | Kind |
|---|---|---|---|
| 1152917 | Apr 2011 | FR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2012/053192 | 2/24/2012 | WO | 00 | 10/4/2013 |
